Nitrogen isotope systematics and origins of mixed-habit diamonds

https://doi.org/10.1016/j.gca.2015.01.033Get rights and content

Abstract

Nitrogen isotope values from mantle diamonds are a commonly used tracer in the quest to track volatiles within the Earth’s mantle through deep time. Interpretations of this isotope data are valid so long as stable isotope fractionation processes in the mantle are understood. The fractionation of nitrogen isotopes between {1 1 1} and {1 0 0} growth sectors is well documented for high-pressure high-temperature (HPHT) synthetic diamonds, but there is little data on whether it also occurs in natural mixed-habit diamonds. We present 91 in-situ nitrogen isotope (δ15N) measurements, along with carbon isotope (δ13C) values and nitrogen abundances [N], obtained from three mixed-habit diamonds by secondary ion mass spectrometry (SIMS). While the well-documented enrichment of nitrogen concentrations in octahedral sectors compared to contemporaneous cuboid sectors is observed, a similarly clear disparity is not obvious in the δ15N data. Whereas HPHT synthetic diamonds exhibit 15N enrichment in the {1 0 0} sectors by ∼+30‰, the mixed-habit diamonds studied here show enrichment of the octahedral sectors in 15N by only 0.4–1‰. This major difference between HPHT synthetic and natural mixed-habit diamonds is proposed to be the result of different physical properties of the growth interfaces. The smooth interfaces of the octahedral sectors are the same in both types of crystal, but the outermost atoms on the smooth cube interfaces of an HPHT synthetic diamond behave differently to those on the rough cuboid interfaces of the natural mixed-habit diamonds, resulting in different δ15N values. Both the δ13C (average of ∼−8.7‰) and δ15N (average of ∼0‰) data show only minor offsets from the typical mantle values (δ13C = −5 ± 3‰, δ15N = −5 ± 4‰). This may indicate diamond formation from a mantle derived fluid/melt containing a minor subducted component (lowering δ13C values and elevating δ15N) or relate to moderate degrees of isotopic fractionation of a pure mantle fluid/melt by prior diamond precipitation. The homogeneous nature of both the carbon and nitrogen isotopic compositions of all three diamonds, however, documents continuous and unlimited supply of diamond forming fluid/melt, with a constant composition. Such homogenous isotopic compositions exclude fluid mixing or isotopic fractionation close to the site of diamond formation and preclude distinguishing between these two processes based on diamond analyses alone.

Introduction

Nitrogen is an important component of all of Earth’s major reservoirs, being present in the biosphere, atmosphere, crust and mantle and predicted in the core, but there are very large differences between these reservoirs in terms of their nitrogen concentrations and isotopic compositions (Busigny and Bedout, 2013). It is these differences and how nitrogen cycles between the reservoirs that gives nitrogen the potential to be an important geochemical tracer (e.g. Halama et al., 2010). However, before nitrogen isotope data can be used effectively, its behavior in the various environments needs to be better understood. Much of our understanding of nitrogen in the mantle comes from the investigation of cratonic diamonds. As nitrogen is the most common impurity in diamond, it can compliment carbon isotope data in tracing the deep carbon cycle through time (Javoy et al., 1984, Boyd et al., 1987, Boyd et al., 1992, Boyd and Pillinger, 1994, Cartigny et al., 1997, Cartigny et al., 1998a, Cartigny et al., 1998b, Cartigny et al., 2001, Cartigny et al., 2003, Cartigny et al., 2004, Cartigny et al., 2009, Harte et al., 1999, Bulanova et al., 2002, Bulanova et al., 2014, Hauri et al., 2002, Gautheron et al., 2005, Klein-BenDavid et al., 2010, Smit et al., 2014, Thomassot et al., 2007, Thomassot et al., 2009, Palot et al., 2009, Palot et al., 2012, Mikhail et al., 2013, Mikhail et al., 2014).

Nitrogen concentrations in diamonds range from 0 to >5000 ppm, and in general nitrogen is strongly enriched in diamonds relative to primitive mantle concentrations (Stachel and Harris, 2009). The mean nitrogen isotope value (δ15N) for peridotitic diamonds is −5 ± 4‰ (Cartigny et al., 2014), however, the range reported for all cratonic diamonds is very large (∼−40‰ to ∼+20‰; Cartigny and Marty, 2013), and the variations seen in individual samples can also be as much as 30‰ (Mikhail et al., 2014). Before assuming these data reflect δ15N heterogeneity in the mantle, constraints on the behavior of nitrogen during diamond formation (e.g. compatibility, isotopic fractionation between growth sectors within the diamond, isotopic fractionation between the diamond and fluid/melt growth medium) need to be better understood. To provide further insights into some of these problems, this study will determine the nitrogen isotope systematics of mixed-habit diamonds.

Mixed-habit diamonds are examples of crystals that form by two growth mechanisms occurring at the same time (Frank, 1967); namely octahedral and cuboid growth (Moore and Lang, 1972, Lang, 1974, Suzuki and Lang, 1976a, Suzuki and Lang, 1976b). Octahedral growth produces the flat {1 1 1} faces that define the iconic diamond-crystal morphology, while cuboid growth produces hummocky, non-faceted surfaces whose mean orientation is {1 0 0}, but can be inclined by up to 30°. Crystal-growth theory of diamond (see review by Sunagawa (2005)) shows that the {1 1 1} faces are smooth interfaces on which layer or spiral growth is expected to occur, while the {1 0 0} faces are rough interfaces on which growth occurs by an adhesive-type mechanism. Although rough faces would be expected to grow much more rapidly than smooth ones and therefore grow themselves out of the crystal (Moore, 1985), it is thought that solute–solvent interaction and/or impurity absorption can reduce the growth rate of the rough crystal faces (Palyanov et al., 2010). This prevents them from growing out of the crystal, and allows them to remain as large as the smooth octahedral, habit-controlling faces (Sunagawa, 2005).

Having two different growth mechanisms occurring within a single crystal produces different impurity characteristics between the two growth sectors. Nitrogen concentrations in mixed habit diamonds are often very high (900 to >2000 ppm) and N is preferentially partitioned into the octahedral sectors; with enrichment ranging from 107% to 157% relative to the cuboid sectors (Welbourn et al., 1989, Lang et al., 2004, Rondeau et al., 2004, Zedgenizov and Harte, 2004, Howell et al., 2012, Howell et al., 2013a). While nitrogen-aggregation levels can vary from 0% to 100% IaB, there seems to be no unanimity in the literature regarding differences between the sectors. However, two recent studies (Howell et al., 2012, Howell et al., 2013a) have shown compelling evidence that the rate of nitrogen aggregation does not differ between sectors, but that the subsequent rate of platelet formation can be retarded in cuboid sectors due to the presence of “disc-crack-like” defects (Walmsley et al., 1987). Hydrogen-related absorption features are significantly stronger in the cuboid sectors (Rondeau et al., 2004, Howell et al., 2012, Howell et al., 2013a), most likely due to C–H bonding occurring on the internal surfaces of the disc-crack-like defects. However, due to the possible abundance of IR-inactive hydrogen within a diamond (Sellschop, 1992, Sweeney et al., 1999), there remains no quantitative evidence regarding hydrogen partitioning between growth sectors. Spectroscopic studies have revealed evidence of nickel-related defects in the cuboid sectors (Welbourn et al., 1989, Lang et al., 2007, Howell et al., 2013a). Subsequent laser-ablation inductively-coupled-plasma mass spectrometry (LA-ICPMS) analysis of mixed-habit diamonds (Howell et al., 2013b) confirmed the preferential partitioning of Ni, as well as Co, into the cuboid sectors. Ni concentrations ranged from 3.6 to 36 ppm, and Co from 0.1 to 1.3 ppm, with enrichment relative to the octahedral sectors ranging from 10 to 460 times and 10 to 100 times, respectively. Carbon-isotope analysis via both gas-sourced mass spectrometry (Cartigny et al., 2003) and secondary-ion mass spectrometry (SIMS; Bulanova et al., 2002, Zedgenizov and Harte, 2004, Howell et al., 2012, Howell et al., 2013a) has revealed no evidence of isotopic fractionation between the growth sectors.

The same partitioning of nitrogen into {1 1 1} sectors is seen in high-pressure high-temperature (HPHT) synthetic cubo-octahedral diamonds, but this is accompanied by significant kinetic nitrogen isotopic fractionation (∼30‰; Boyd et al., 1988). To date, two studies have investigated this nitrogen isotopic fractionation in natural mixed-habit diamonds. The first study used SIMS on a mixed-habit diamond from the Mir kimberlite (Yakutia, Russia; Bulanova et al., 2002). While no fractionation between the two growth sectors was reported, this was not the primary focus of the study. The diamond had δ15N values of ∼−3‰ for the bulk of the crystal (core and intermediate zones) with a few values of +8.6‰ to +9.3‰ in the rim zone. However, it should be noted that δ15N analyses via SIMS at that time had an analytical precision of ±2‰, and accuracy of ±8‰ (2σ; Hauri et al., 2002). This means that any isotopic fractionation <2‰ could not have been detected.

The second study of nitrogen isotopes in a mixed-habit diamond was by Cartigny et al. (2003), which used gas-sourced mass spectrometry following the combustion of the sample. While that study also found no nitrogen-isotope fractionation (a 0.1‰ difference between growth sectors with 0.5‰ analytical uncertainties), our interpretation of the cathodoluminescence (CL) image of the sample used (Fig. 1 in Cartigny et al. (2003)), is that this sample was not actually of mixed habit. The CL reveals smooth flat growth in the “cubic” sector, as opposed to the hummocky/curved layers that are characteristic of cuboid growth. It is possible the diamond was only bound by {1 1 1} faces but had a malformed habit, making it more tetrahedral in appearance that was mistaken for cubo-octahedral (Sunagawa, 2010). While the sample did show evidence of moderate nitrogen partitioning between the two sectors, 23–29 ppm in the {1 0 0}, 42–57 ppm in the {1 1 1}, it does not exhibit the large overall nitrogen concentrations and differences between sectors seen in all other mixed-habit diamonds (see above). Combining these two observations leads us to conclude that this sample did not contain any true cuboid growth, and as such the study’s conclusion that no fractionation takes place in natural mixed-habit diamonds is unsubstantiated. As a result, there is clearly a lack of detailed nitrogen-isotope data from mixed-habit diamonds available in the literature to discern whether the preferential partitioning of nitrogen in {1 1 1} growth sectors results in significant isotopic fractionation in the mantle. Thus, this process may or may not result in large isotopic excursions through solely intra-mantle processes.

Here we present N-isotope and N-concentration data, as well as C-isotope data, from three mixed-habit diamonds obtained via SIMS, with improved analytical precision (0.5–0.9‰) and accuracy (<1.5‰) compared to earlier studies (Bulanova et al., 2002, Hauri et al., 2002). Firstly, we investigate whether nitrogen isotopes show any preferential fractionation between growth sectors and compare these findings with those from HPHT synthetic diamonds. Then we use these nitrogen-isotope data to further refine our understanding of how these unusual diamond crystals form, and to better constrain the behavior of nitrogen during diamond formation in the mantle.

Section snippets

Samples

Three mixed-habit diamonds were analyzed in this study (Fig. 1); they were taken from the collection analyzed by Howell et al., 2013a, Howell et al., 2013b. As the samples were obtained from the gem industry, their geographical source is unknown. They are all in the form of doubly polished {1 0 0} plates, ranging in size from 7 to 10 mm (longest axis), all less than 1 mm thick. The previously obtained spectroscopic and carbon isotope data (Howell et al., 2013a) and LA-ICPMS data (Howell et al.,

Results

In total, 134 carbon-isotope and 91 nitrogen-isotope and N-concentration measurements were made on the three samples. All of the individual analyses discussed below are provided in the Supplementary Data Table, with a summary of each sample given in Table 1.

Analyses (n = 46) of both carbon and nitrogen isotopes were obtained from transects of MC08B (transect XS2, n = 20) and MC08C (transect XS1, n = 26), Fig. 1, Fig. 3. In both transects, the analyses were performed at ∼100 μm intervals, with the

Discussion

While the focus of this study is the new nitrogen isotope data from these mixed-habit diamond samples (discussed below), it is important to first make two observations on the reproducibility of the carbon-isotope and nitrogen-concentration data. Firstly, the C-isotope data recorded from MC08 in this study are in complete agreement with the previously obtained data from this sample (Howell et al., 2013a). Also, the ranges of C-isotope data obtained from MC10 (δ13C = −8.37‰ to −8.79‰) and MC13 (δ13C

Conclusions

The new SIMS methodology for δ15N analysis presented in this study has greatly improved the data quality (both accuracy and precision) for this in-situ technique. It is now at the point where it can provide valuable insights into the stable-isotope systematics of individual diamonds. The homogeneous δ13C and δ15N data obtained for three mixed-habit diamonds in this study suggest they grew from a constant homogenous medium in an open (non-fluid limited) system. In the absence of internal

Acknowledgements

Pierre Cartigny, Ben Harte and an anonymous reviewer are thanked for their constructive comments that helped to improve this manuscript, and Marc Norman is thanked for his editorial handling. Tom Chacko is thanked for helpful discussions. RS would like to thank EPSRC and the DTC for provision of a PhD studentship. The CCIM acknowledges operational funding from the Canada Foundation for Innovation, Alberta Innovates, and the University of Alberta. This contribution relates to CCIM project P1301.

References (69)

  • T. Chacko et al.

    Oxygen and carbon isotope fractionations between CO2 and calcite

    Geochim. Cosmochim. Acta

    (1991)
  • P. Deines

    The carbon isotopic composition of diamonds: relationship to diamond shape, color, occurrence and vapor composition

    Geochim. Cosmochim. Acta

    (1980)
  • P. Deines et al.

    Nitrogen and 13C content of Finsch and Premier diamonds and their implications

    Geochim. Cosmochim. Acta

    (1989)
  • C. Gautheron et al.

    Evidence for a mantle component shown by rare gases, C and N isotopes in polycrystalline diamonds from Orapa (Botswana)

    Earth Planet. Sci. Lett.

    (2005)
  • R. Halama et al.

    Nitrogen recycling in subducted oceanic lithosphere: the record in high- and ultrahigh-pressure metabasaltic rocks

    Geochim. Cosmochim. Acta

    (2010)
  • M.S. Hamza et al.

    Surface effect on the isotopic fractionation between CO2 and some carbonate minerals

    Geochim. Cosmochim. Acta

    (1974)
  • E.H. Hauri et al.

    Microanalysis of δ13C, δ15N, and N abundances in diamonds by secondary ion mass spectrometry

    Chem. Geol.

    (2002)
  • D. Howell et al.

    Trace element partitioning in mixed-habit diamonds

    Chem. Geol.

    (2013)
  • M. Javoy et al.

    Nitrogen and carbon isotopic composition in the diamonds of Mbuji Mayi (Zaire)

    Earth Planet. Sci. Lett.

    (1984)
  • O. Klein-BenDavid et al.

    Mixed fluid sources involved in diamond growth constrained by Sr–Nd–Pb–C–N isotopes and trace elements

    Earth Planet. Sci. Lett.

    (2010)
  • K.T. Koga et al.

    Diffusive relaxation of carbon and nitrogen isotope heterogeneity in diamond: a new thermochronometer

    Phys. Earth Planet. Inter.

    (2003)
  • A.R. Lang et al.

    Is dispersed nickel in natural diamonds associated with cuboid growth sectors in diamonds that exhibit a history of mixed-habit growth?

    J. Cryst. Growth

    (2004)
  • A.R. Lang et al.

    Defects in a mixed-habit Yakutian diamond: studies by optical and cathodoluminescence microscopy, infrared absorption, Raman scattering and photoluminescence spectroscopy

    J. Cryst. Growth

    (2007)
  • Y. Li et al.

    Nitrogen speciation in mantle and crustal fluids

    Geochim. Cosmochim. Acta

    (2014)
  • S. Mikhail et al.

    Peridotitic and websteritic diamondites provide new information regarding mantle melting and metasomatism induced through the subduction of crustal volatiles

    Geochim. Cosmochim. Acta

    (2013)
  • S. Mikhail et al.

    Constraining the internal variability of the stable isotopes of carbon and nitrogen within mantle diamonds

    Chem. Geol.

    (2014)
  • M. Palot et al.

    Diamond origin and genesis: A C and N stable isotope study on diamonds from a single eclogitic xenolith (Kaalvallei, South Africa)

    Lithos

    (2009)
  • M. Palot et al.

    Evidence for deep mantle convection and primordial heterogeneity from nitrogen and carbon stable isotopes in diamond

    Earth Planet. Sci. Lett.

    (2012)
  • M. Palot et al.

    Multiple growth events, processes and fluid sources involved in diamond genesis: a micro-analytical study of sulphide-bearing diamonds from Finsch mine, RSA

    Geochim. Cosmochim. Acta

    (2013)
  • M. Palot et al.

    Isotopic constraints on the nature and circulation of deep mantle C–H–O–N fluids: carbon and nitrogen systematics within ultra-deep diamonds from Kankan (Guinea)

    Geochim. Cosmochim. Acta

    (2014)
  • V.B. Polyakov et al.

    The use of heat capacity data to calculate carbon isotope fractionation between graphite, diamond, and carbon dioxide: a new approach

    Geochim. Cosmochim. Acta

    (1995)
  • B. Rondeau et al.

    Three historical ‘asteriated’ hydrogen-rich diamonds: growth history and sector-dependent impurity incorporation

    Diam. Relat. Mater.

    (2004)
  • K.A. Smart et al.

    Diamond growth from oxidized carbon sources beneath the Northern Slave Craton, Canada: a δ13C–N study of eclogite-hosted diamonds from the Jericho kimberlite

    Geochim. Cosmochim. Acta

    (2011)
  • T. Stachel et al.

    Sources of carbon in inclusion bearing diamonds

    Lithos

    (2009)
  • Cited by (18)

    • Effect of growth rate on diamond composition

      2023, Diamond and Related Materials
    • Kinetic carbon isotope fractionation links graphite and diamond precipitation to reduced fluid sources

      2020, Earth and Planetary Science Letters
      Citation Excerpt :

      A kinetic control on carbon and nitrogen isotope ratios in synthetic and natural diamonds was already discussed as a possible mechanism by Boyd and coworkers (1988, 1994), Deines (2002), and Bulanova et al. (2002). Only a few coherent diamond growth zones have been found to show well-resolved isotope trends, often, variations are within error of the analytical technique (e.g. Howell et al., 2015; Smit et al., 2016, 2019). Their magnitude of carbon isotope fractionation in coherent growth zones appears small (<1.3 ‰, Smit et al., 2016; Smart et al., 2011) compared to profiles along oscillatory zonings that may span several permil (e.g. >2.5 ‰ in Petts et al., 2015).

    • Analysis of an unusual mixed-habit natural diamond by high spatial resolution techniques: Growth and fracture

      2018, Journal of Crystal Growth
      Citation Excerpt :

      As a way of investigating the reliability of FTIR measurements in determining the total nitrogen content it is interesting to refer to examples where FTIR results are compared with other techniques that give the total nitrogen concentration. In recent work by Howell et al. [30] FTIR and SIMS results were obtained for three diamonds, with clearly distinguished octahedral and cuboidal regions. The agreement between the two techniques was quite close, the SIMS values being 2–15% larger.

    • Diamond growth from C–H–N–O recycled fluids in the lithosphere: Evidence from CH<inf>4</inf> micro-inclusions and δ<sup>13</sup>C–δ<sup>15</sup>N–N content in Marange mixed-habit diamonds

      2016, Lithos
      Citation Excerpt :

      In this case, appropriate sector-dependent source-diamond fractionation factors would be needed to determine source compositions. The similarity of carbon and nitrogen isotopes in octahedral and cuboid sectors of natural mixed-habit diamonds has been well documented (Bulanova et al., 2002; Howell et al., 2013a, 2015; Zedgenizov and Harte, 2004), and in Marange diamonds as well, no isotopic differences between cuboid and octahedral growth sectors are observed. Therefore, overall variation in isotopic compositions through the suite (see discussion below), regardless of growth sector, can be used to assess growth conditions and source compositions.

    • Morphology of Monocrystalline Diamond and its Inclusions

      2023, Diamond: Genesis, Mineralogy and Geochemistry
    View all citing articles on Scopus
    View full text